Validity of Recogniton of Stroke in the Emergency Department (ROSIER) scale in Diagnosis of Stroke in Emergency Department

Abstract

Determining the external diagnostic value of diagnostic criteria is important in clinical trials in different places. According to the literature review, no study has been conducted to determine the diagnostic value of the ROSIER scale in Iran and the northwest of the country. On the other hand, the incidence of stroke in our country is higher than in Western countries, and as it is clear, if a stroke is diagnosed quickly in the emergency, it can take the necessary measures for faster interventions. Therefore, the aim of the present study was to determine the diagnostic value of the ROSIER criterion in the diagnosis of emergency stroke.

Methods: In this cross-sectional study, patients who referred to the emergency department of Imam Reza (AS) Hospital in Tabriz from June to September 2020 and were suspected of having a stroke were included in the study according to the inclusion and exclusion criteria. At the beginning of the visit, after the patient stabilized, the ROSIER standard score for patients was calculated by an emergency medicine specialist. In addition to the ROSIER criterion, patients were evaluated as the Gold Standard by the NIHSS criterion. Patient information including age, sex, time of onset of symptoms, time of emergency visit, time between stroke and emergency room visit, clinical signs and symptoms, systolic and diastolic blood pressure, heart rate, body temperature, number of breaths, score (Glasgow Coma Score (GCS), history of underlying (previous) illness, imaging findings, final diagnosis, initial outcome will be recorded at discharge.

Results: In the ROC curve, the level below the graph for ROSIER with a diagnosis of 0.763, which indicates the relatively good usefulness of the test, and therefore the sensitivity is equal to 96.49%, while the specificity is equal to 5.56%. The positive predictive value of this test is 90.66% and its negative predictive value is 14.29%. Therefore, with this test, a stroke can be diagnosed with high confidence, but a healthy person with the same symptoms as a stroke cannot be isolated and can not rule out a stroke.